JP2005532585A - Audio coding - Google Patents

Abstract

The encoding (1) of the audio signal (x) comprises providing a respective set of sampled signal values for each of a plurality of consecutive segments. The sampled signal value is analyzed 130 and one or more sinusoidal components are generated for each of the plurality of consecutive segments. A sinusoidal component is linked between a plurality of consecutive segments. A sinusoidal code (CS) has a track of linked sinusoidal components for each of the plurality of consecutive segments. Each of the tracks has a frequency and amplitude of a sinusoidal component in the starting segment of the track, while the selected track includes an indicator that the phase of the starting segment is not included.

Description

  The present invention relates to encoding and decoding of audio signals.

  Parametric coding schemes, particularly sinusoidal coders, are described in International Patent Application No. WO 00 / 79519-A1 (reference number PHN017502) and PCT Patent Application No. IB02 / 01297 (reference number PHNL010252). In this coder, an audio segment or frame is modeled by a sine wave coder with a number of sine waves represented by amplitude, frequency and phase parameters. Once the sine wave of the segment is estimated, the tracking algorithm is started. This algorithm attempts to link sine waves together segment by segment. In this way, sine wave parameters from the appropriate sine wave from successive segments are linked to obtain a so-called track. The link criteria is based on the frequency of two consecutive segments, but also amplitude and / or phase information can be used. This information is combined in a cost function that determines the sine wave to be linked. Thus, the tracking algorithm produces a sinusoidal track that starts at a particular time instance, develops for a period of time over multiple time segments, and stops.

  In this scheme, the initial phase is transmitted for a sine wave track, and the phase of the other sine waves in the track is obtained from this initial phase and the frequency of these other sine waves. The amplitude and frequency of the sine wave can also be encoded differently than the previous sine wave. Furthermore, very short tracks can be removed.

  Thus, due to tracking, the bit rate of the sine wave coder can be significantly reduced.

  According to the invention, there is provided a method for encoding an audio signal according to claim 1.

  In the preferred embodiment of the present invention (FIG. 1), the encoder is a sine wave coder of the type described in WO 01 / 69593-A1 (reference number PHNL000120). The operation of this coder and its corresponding decoder is well described, and only a description of the parts relevant to the present invention is given here.

  In both the previous case and the preferred embodiment, the audio coder 1 samples the input audio signal at a specific sampling frequency, resulting in a digital representation x (t) of the audio signal. The coder 1 then separates the sampled input signal into three components: a transient signal component, a persistent deterministic component and a persistent stochastic component. The audio coder 1 includes a transient coder 11, a sine wave coder 13, and a noise coder 14. The audio coder optionally has a gain compression mechanism (GC) 12.

  The transient coder 11 includes a transient detector (TD) 110, a transient analyzer (TA) 111, and a transient synthesizer (TS) 112. First, the signal x (t) enters the transient detector 110. The detector 110 estimates whether there is a transient signal component and the position of the signal component. This information is supplied to the transient analyzer 111. If the position of the transient signal component is determined, the transient analyzer 111 attempts to extract the (main part) of the transient signal component. The analyzer aligns the shape function with a signal segment, preferably starting at the estimated starting position, and determines the content under the shape function using, for example, a (small) number of sinusoidal components. This information is included in the transient code CT. More detailed information for generating the transient code CT is given in International Patent Application No. 01 / 69593-A1.

  The transient code CT is supplied to the transient synthesizer 112. The synthesized transient signal component is subtracted from the input signal x (t) in the subtracter 16, and as a result, the signal x1 is obtained. When GC 12 is omitted, x1 = x2.

  The signal x2 is fed to a sine wave coder 13 and analyzed in a sine wave analyzer (130), which determines the (deterministic) sine wave component. Thus, although the presence of a transient analyzer is desirable, this is not essential and it will be appreciated that the present invention can be implemented without such an analyzer. In any case, the result of sinusoidal coding is a sinusoidal code CS, and a more detailed example showing the conventional generation of an exemplary sinusoidal code CS is given in WO 00 / 79519-A1. It is done.

  In essence, however, such a sine wave coder encodes the input signal x2 as a track of a sine wave component that is linked from one frame segment to the next. In the prior art, a track is first represented by a starting frequency, starting amplitude and starting phase for a sine wave starting in a given segment (birth).

  In the preferred embodiment of the invention, the starting phase is selectively encoded as a function of the length of the track. More specifically, the start phase is only used for long duration tracks. This is because a long-lived track is probably encoded with tonal information, and in such a case, the acoustic characteristics of the track are as much as possible by transmitting the starting phase of the track. This is because it is presumed that it is important to preserve. Short-lived tracks are presumed to encode non-tonal information, so transmitting a start phase with such a track is actually a sonic property to the track. Is thus estimated to give a perception of distortion when playing the encoded bitstream.

  It will be appreciated that by not transmitting the start phase for short tracks, significant bit rate savings can occur. This is because the overhead of starting phase data for short tracks is proportionally higher than for longer tracks.

  There are many alternative criteria for determining whether a track is long enough to require a starting phase or correspondingly short enough not to require a starting phase.

  The simplest criterion is to select an absolute track length. Experimentally, it has been discovered that tracks shorter than 40 ms do not require a start phase, while longer tracks are advantageously transmitted with a start phase. In an encoder with an update interval of 8 ms, this means that a track that is less than 5 segments in length does not contain a start phase, but an indicator that the start phase is not used by that track. . (Encoding such an indicator is presumed to be more efficient compared to the starting phase value.) Alternatively, the encoded signal produced by the encoder is a compatible decoder. The encoder does not have to include an indication that the starting phase is not used, but can leave the decoder to determine how to process the track without the starting phase. .

The alternative criteria is based on determining whether the time interval in which a track is present is voiced or non-voiced. If the time interval is determined to be voiced, this effectively silent time interval (and thus the track) should not include the start phase, and vice versa for unvoiced time intervals, It is estimated to be. LR
Rabiner, MJ Cheng, AE Rosenberg and CA
“A” by McGonegal
Comparative Performance Study of Several Pitch Detection Algorithms '' (IEEE
Transactions on Acoustics, Speech and Signal Processing, ASSP-24, pp. 399-417, October 1976) discloses a method for making such a determination, and in this method, components that realize such a method. Is included in the tracking algorithm so that the tracking algorithm includes start phase information for tracks that exist within a sound time interval, while for tracks that exist within a non-sound time interval, The starting phase is not included in the encoded bitstream. This criterion is that in a sound time interval, the track tends to be longer than in a non-sound time interval, so the final of the track is determined before a determination is made whether the track should contain a start phase. Suppose the specific length need not be known.

  An alternative way of determining whether the time interval represents a sonic or non-audio audio signal is to examine the energy level of the noise component of the signal, as described below. Once the ratio of noise energy to sinusoidal component energy is found to exceed a given threshold for a given time interval, the audio signal is silent and the start phase information is included in the track, as above. It can be assumed that there is no need. The reverse is also true when the ratio of noise energy to sinusoidal component energy falls below a given threshold. Again, if the signal is determined to be sonic, the track is assumed to be longer than in the non-phonic signal.

  In both the preferred embodiment and the prior art, a track is represented by a continuous segment by frequency and amplitude differences (or by phase difference (continuation) for longer tracks) until the track ends (dead) segment. . In practice, it can be determined that there is little gain in encoding the phase difference even for long tracks. Thus, the phase information need not be encoded for continuation, and the phase information for long tracks can be regenerated using continuous phase recovery.

  Similar to the prior art, the sine wave signal component is recovered by the sine wave synthesizer (SS) 131 from the sine wave code CS generated by the improved sine wave coder of the present invention. This signal is subtracted from the input x2 to the sine wave coder 13 in a subtractor 17, resulting in a residual signal x3 that does not contain a (large) transition signal component and a (main) deterministic sine wave component.

  The residual signal x3 is presumed to have mainly noise, and the noise analyzer 14 of the preferred embodiment is a noise that is an indication of this noise, as described, for example, in International Patent Application No. 01 / 89086-A1. The code CN is generated. Again, it will be appreciated that the use of such an analyzer is not essential for the realization of the present invention and is complementary to such use.

  Finally, the multiplexer 15 forms an audio stream AS including codes CT, CS, and CN. The audio stream AS is supplied to, for example, a data bus, an antenna system, a storage medium, and the like.

  FIG. 2 shows an audio player 3 according to the invention. For example, the audio stream AS ′ generated by the encoder according to FIG. 1 is obtained from a data bus, an antenna system, a storage medium or the like. The audio stream AS is demultiplexed by the demultiplexer 30, and codes CT, CS, and CN are obtained. These codes are supplied to the transient synthesizer 31, the sine wave synthesizer 32, and the noise synthesizer 33, respectively. The transient signal component is calculated from the transient code CT in the transient synthesizer 31. If the transient code indicates a shape function, the shape is calculated based on the received parameters. Further, the shape content is calculated based on the frequency and amplitude of the sine wave component. If the transient code CT indicates a step, no transient is calculated. The total transient signal yT is the sum of all transients.

  A sine wave code CS is used to generate the signal yS, which is described as the sum of the sine waves of a given segment. At the decoder, the phase of the sine wave of the sine wave track is determined in one of two ways. If the track contains a starting phase, the phase is calculated from the resulting sine wave phase and the intermediate sine wave frequency, as in the prior art. In a preferred embodiment where the track includes an indication that no start phase is provided, the decoder generates a random start phase for all sine waves of the track and synthesizes the track as before. (The decoder may alternatively calculate a random starting phase only for the resulting sine wave, and the remaining phases may be calculated as in the prior art.) No such indication or starting phase is provided. In some cases, the decoder assumes that it is required to create a random starting phase for the sine wave of the track.

  It will be appreciated that one aspect of the invention is to preserve the non-tonality of non-audio audio fragments. Thus, when utilizing the present invention, it may be desirable for the encoder to store very short tracks for non-audio audio fragments and for the decoder to play these short tracks with a random starting phase. is there. This is different from the prior art where very short tracks are not included anywhere in the bitstream.

  At the same time, the noise code CN is supplied to the noise synthesizer NS33, which is mainly a filter having a frequency response approximating the noise spectrum. The NS 33 generates a noise yN restored by filtering the white noise signal by the noise code CN.

The total signal y (t) has the sum of the product of the transient signal yT and any amplitude expansion (g) and the sum of the sinusoidal signal yS and the noise signal yN. The audio player has two adders 36 and 37 for taking the sum of the respective signals.
The total signal is supplied to the output unit 35, which is for example a speaker.

  FIG. 3 shows an audio system according to the invention having the audio coder 1 shown in FIG. 1 and the audio player 3 shown in FIG. Such a system provides playback and recording functions. The audio stream AS is supplied from the audio coder to the audio player through the communication channel 2, which may be a wireless connection, a data 20 bus or a storage medium. When the communication channel 2 is a storage medium, the storage medium may be fixed to the system, or may be a removable disk, memory stick, or the like. The communication channel 2 may be part of the audio system, but is often external to the audio system.

  The present invention can be used in any sinusoidal audio coder. Thus, the present invention is applicable wherever such a coder is used.

  It should be noted that the above-described embodiments illustrate rather than limit the invention, and that those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word “comprising” does not exclude the presence of elements or steps other than those listed in a claim. The present invention can be implemented by hardware having several individual elements and by a suitably programmed computer. In the device claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage.

1 shows an embodiment of an audio coder according to the present invention. 1 shows an embodiment of an audio player according to the present invention. 1 shows a system having an audio coder and an audio player according to the present invention.

Claims (15)

In a method of encoding an audio signal,
Providing a respective set of sampled signal values for each of a plurality of consecutive segments;
Analyzing the sampled signal value to generate one or more sinusoidal components for each of the plurality of consecutive segments;
Linking sinusoidal components between a plurality of consecutive segments;
Generating a sine wave code having linked sine wave component tracks for each of the plurality of consecutive segments, each of the tracks having a frequency and amplitude of a sine wave component of a start segment of the track; And the selected track does not include the phase of the start segment; and
Generating an encoded audio stream including the sine wave code;
Having a method.   2. The method of claim 1, wherein the selected track includes an indicator that the phase of the starting segment is not included.   The method of claim 1, wherein the selected track is less than 5 segments in length.   The method of claim 1, wherein the selected track is less than 40 ms in length.   The method of claim 1, wherein the selected track represents a silent component of an audio signal.   The method of claim 1, wherein the selected track represents a component of a voiced time interval of the audio signal.   The method of claim 1, wherein the selected track represents a noisy interval component of the audio signal.   The method of claim 1, wherein each of the tracks has a frequency and amplitude difference for each sinusoidal component of the next continuation segment of the track. In a method for decoding an audio stream,
Reading an encoded audio stream including a sinusoidal code having a track of linked sinusoidal components for each of a plurality of consecutive segments, each of the tracks being a sinusoid of a start segment of the track Having a frequency and amplitude of components, and the selected track does not include the phase of the start segment;
Generating a random starting phase for the selected track;
Synthesizing the audio signal using the sine wave code comprising restoring a sine wave component over a plurality of consecutive segments;
Having a method.   10. The method of claim 9, wherein the generating step includes generating a random phase for each sine wave component of the selected track. In an audio coder configured to process a respective set of sampled signal values for each of a plurality of consecutive segments of an audio signal,
An analyzer configured to analyze the sampled signal value to generate one or more sinusoidal components for each of the plurality of consecutive segments;
A linker configured to link sinusoidal components between a plurality of consecutive segments;
A component configured to generate, for each of the plurality of consecutive segments, a sine wave code having a track of linked sine wave components, each of the tracks being a sine wave component of a start segment of the track The selected track does not include the phase of the start segment; and
A bitstream generator for generating an encoded audio stream including the sine wave code;
Audio coder with. Means for reading an encoded audio stream comprising a sinusoidal code having a track of linked sinusoidal components for each of a plurality of consecutive segments, each of said tracks comprising a start segment of the track Means having a frequency and amplitude of a sinusoidal component, wherein the selected track does not include the phase of the start segment;
A phase generator configured to generate a random starting phase for the selected track;
A synthesizer for synthesizing the audio signal using the sine wave code, comprising restoring a sine wave component over a plurality of consecutive segments;
An audio player.   An audio system comprising the audio coder according to claim 11 and the audio player according to claim 12.   In an audio stream having a sinusoidal code representing at least a component of the audio signal, the code has a track of sinusoidal components linked between the plurality of consecutive segments, each of the tracks being a start segment of the track. An audio stream having a frequency and amplitude of a sinusoidal component and the selected track does not include the phase of the start segment.   A storage medium in which the audio stream according to claim 14 is stored.

Images